TNG: DeepSeek R1T Chimera

Text input Text output Free Option
Author's Description

DeepSeek-R1T-Chimera is created by merging DeepSeek-R1 and DeepSeek-V3 (0324), combining the reasoning capabilities of R1 with the token efficiency improvements of V3. It is based on a DeepSeek-MoE Transformer architecture and is optimized for general text generation tasks. The model merges pretrained weights from both source models to balance performance across reasoning, efficiency, and instruction-following tasks. It is released under the MIT license and intended for research and commercial use.

Key Specifications
Cost
$$$
Context
163K
Parameters
1T
Released
Apr 27, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Reasoning Include Reasoning Seed Top P Temperature Top Logprobs Logit Bias Logprobs Stop Min P Max Tokens Frequency Penalty Presence Penalty
Features

This model supports the following features:

Reasoning
Performance Summary

TNG: DeepSeek R1T Chimera, a merge of DeepSeek-R1 and DeepSeek-V3 (0324), demonstrates exceptional speed, consistently ranking among the fastest models with an Infinityth percentile across 8 benchmarks. It offers competitive pricing, placing in the 47th percentile across 7 benchmarks. The model exhibits outstanding reliability, achieving a 100% success rate across all 8 benchmarks, indicating minimal technical failures. In terms of performance across categories, DeepSeek R1T Chimera shows remarkable strengths in General Knowledge and Ethics, achieving perfect 100% accuracy in both, and is noted as the most accurate model at its price point and among models of similar speed. Its Coding capabilities are also strong, with 95.0% accuracy (97th percentile), and Reasoning is highly proficient at 98.0% accuracy (94th percentile). While its Hallucinations accuracy is 94.0%, indicating a good ability to acknowledge uncertainty, its Instruction Following performance is a notable weakness, with a 1.0% accuracy in one benchmark and 0.0% in another, suggesting significant challenges with complex multi-step instructions. Email Classification is solid at 98.0% accuracy. Overall, the model excels in knowledge-based and ethical reasoning tasks, but requires improvement in instruction adherence.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.25
Completion $1

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Chutes
Chutes | tngtech/deepseek-r1t-chimera 163K $0.25 / 1M tokens $1 / 1M tokens
Chutes
Chutes | tngtech/deepseek-r1t-chimera 163K $0.25 / 1M tokens $1 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by tngtech